JPS59188390A - Controller of thyristor converter - Google Patents

Abstract

PURPOSE:To improve the response by detecting a load current which becomes from an interrupted state to a continuous state and matching the constant of a main circuit to that of an actual main circuit. CONSTITUTION:An arccosine converter 10 outputs a set control angle (alpha) from a phase control signal VR, and a complement coefficient calcultor 11 outputs a complement coefficient K1 in response to the angle (alpha). A function generator 14 outputs a deviation angle theta in response to the value produced by multiplying the constant K of a main circuit and a motor current Ia by the coefficient K1. An adder 15 adds the angle (alpha) and the angle theta, and applies a control angle L' to a gate circuit 20. A boundary value passage detector 30 closes a switch 22 at the side (a) at the load current interrupting time and closes at the side (b) at the continuous time. A memory circuit 26 memorizes the output of a multiplier 13 when the load current passes the boundary between the interruption and the continuation. A multiplier 24 multiplies the set value Ka of the main circuit by the output of a divider 22 and outputs a constant K of the main circuit.

Description

[Detailed Description of the Invention] [Field of the Invention] The present invention relates to a control device for a cyrisk converter that can vary the power supplied to a load that generates a back motive force by ignition phase control, and in particular, The present invention relates to a control device for a thyristor converter that compensates for response characteristics due to nonlinearity when current is interrupted.

[Background of the invention]

As is well known, by the thyristor converter! S
l It is practiced to drive a DC motor or an AC motor.

By the way, when a TE motor is driven by a Cyrisk converter, the current flowing through the Cyrisk converter may be continuous or intermittent depending on the load condition. The conversion characteristics of the Cyrisk converter become nonlinear when the current is interrupted. Therefore, as is well known, the response characteristics of the control device for the thyristor converter deteriorate.

To solve this problem, the following nonlinear compensation method has been proposed. This nonlinear compensation method calculates the difference (control deviation angle) in the firing control angle (control delay angle) at which the DC output average voltage of the thyristor converter is the same when the current is continuous and when the current is intermittent, and when the current is intermittent, the phase control signal is Nonlinear compensation is performed by adding the control deviation angle to the phase setting angle found from the above to obtain the ignition control angle.

When this nonlinear compensation method is put into practice, the control deviation angle must be calculated. The control deviation angle must be calculated mainly by considering the main circuit constant determined by the electric motor. Even if a motor is manufactured to specifications, there will be errors in the motor constants, and the motor constants will change over time. If the main circuit constant setting value and the actual main circuit constant do not match, compensation will not be made even if the current is in the intermittent current region, even if it is compensated when the current is continuous.

If the main circuit constant setting value is smaller than the actual main circuit constant,
Even if the current is continuous, it is determined that the current is intermittent and the control deviation angle is added, resulting in overcompensation. In the opposite case, even if the current is in an intermittent state, there will be a region in which it is determined that the current is continuous, and even if it is determined that the sleep flow is in an intermittent state, the control deviation angle to be compensated will be small, resulting in insufficient compensation.

When overcompensation occurs, the DC output voltage of the thyristor converter becomes smaller with respect to the phase control signal. When the DC voltage decreases, the load current decreases, resulting in a positive feedback state in which the control deviation angle to be compensated increases. On the other hand, in the case of insufficient compensation, there will be a region in which nonlinear compensation is not performed even when the current is interrupted, and even if the region is compensated, the control deviation angle will be too small.

In this way, in both cases of over-compensation and under-compensation, it becomes impossible to perform optimal non-linear compensation and improve responsiveness.

The present invention has been made in view of the above-mentioned problems, and its object is to provide a control device for a thyristor converter that can perform optimal nonlinear compensation according to actual main circuit constants.

The feature of the present invention is that it detects the negative Vi current (interruption boundary value) where the load current changes from an intermittent state to a continuous state, and automatically sets the main circuit constant to match the actual main circuit constant. The reason is that nonlinear compensation is performed.

First, a stationary Leonard control device that drives a p-direct current motor using a thyristor converter in which thyristors are connected in a Graetz connection will be described with reference to FIG. 1, which is a typical example of the present invention.

In Figure 1, 1 is a power transformer, 2 is a current transformer that detects alternating current, 3 is a thyristor converter that converts commercial frequency alternating current to variable voltage direct current, 4 is a direct current motor, and 5 is a speed detector. , 6 is a speed control circuit that generates a current command value IR according to the deviation between the speed command value SR and the speed feedback value SF from the speed detector 5, and 7 is a current detection circuit that converts the output of the current transformer 2 into DC. Circuit, 8I'i current command value IR and current detection circuit 7
Control angle command vR according to the deviation from the current feedback value from
A current control circuit 9 generates a firing pulse at a firing control angle α according to a control angle command vR, and a thyristor converter 3
This is a phase control circuit that performs ignition control.

The operation of this configuration is well known and detailed explanation will be omitted, but in short, the thyristor converter 3 is ignited at the control angle α according to the control angle command VR such that the motor current ■1 becomes the current command value In. By performing the control, the stain motive speed SF is controlled to become the speed command value SR.

FIG. 2 shows an embodiment of the present invention.

In FIG. 2, the same symbols as in FIG. Correction coefficient operation $6 to obtain the correction coefficient kl as shown in Figure 3
．． 12 is a multiplier that determines the intermittent start point, which multiplies the main circuit constant determined by the power supply voltage E2 and the inductance X (power supply and electric power unit) by the motor current. 13 is a multiplier that multiplies the correction coefficient kl and the output of the multiplier 12 by ■, 14 is a function generator that outputs the deviation angle θ as shown in FIG. 4 according to the output of the multiplier 13, and 15 is a setting control. An adder that adds the angle α and the deviation angle θ and outputs the control angle α', 20 is a gate output circuit that provides a firing pulse to the thyristor converter 3, 21 is a switch, and 22 is closed to the a side when the current is interrupted. , selector switch that closes to b 1011 when continuous, 24
is a multiplier that multiplies the main circuit constant setting value by the output of the divider 25, and 26 is a memory that stores the output of the diameter calculator 13 (current evaluation signal) when the load current passes the disconnection boundary value. circuit, 27 is an instantaneous current detector, 28 is a zero current detector, 2
9 is a discontinuity detector that detects continuity and discontinuity of load current; 30
is a boundary value passage detector, which detects that the load current has passed through the disconnection boundary 1, and a 32d average current detection circuit.

The operation will be explained next.

In order to facilitate understanding of the present invention, we will explain that nonlinear compensation can be performed by adding the control deviation angle when the current is interrupted.

The relationship between the AC appropriate voltage (secondary voltage of the power transformer 1) E2 applied to the thyristor converter 3 during continuous current, the DC output voltage hd, and the ignition angle α is approximated as follows: can.

, 3v″T Vd= −E2CIJS(t −−
-(1) Also, the relationship between the phase control signal vR input to the phase control circuit 9 and the DC voltage vd is 1.3Vf Vd1,000□E 2 C03V R (2
), it becomes nonlinear.

In order to linearize this, the phase shift characteristic of the phase control circuit 9 is α=
CO3''V it.In other words, equation (2) becomes 3V7 Vd 1,000□E 2CO3 (to O3''''VR).

In this way, even if the relationship between the phase control signal VR and the DC voltage Vd is linearized, the conversion characteristics of the Cyrisk converter 3 become nonlinear.

This will be explained with reference to FIGS. 5 and 6 for a three-phase thyristor converter.

Now, as shown in Figure 5 (a), it is assumed that the ignition control angle is α1 and the motor current {circle around (2)} is conducted intermittently by θ1.

In this state, the DC voltage (instantaneous voltage) Vo of the thyristor converter becomes the induced voltage VM of the DC motor 4 during the non-peeping period (-1 θ!). Therefore, the average voltage V
a 1 v′i It looks like a dotted line that is larger by the amount of hunting.

On the other hand, in order to generate the same average voltage Va when the current is continuous as the average voltage Va when the current is interrupted, the control angle is α2 as shown in FIG. 5(b). As is clear from the figure, the control angle α2
is smaller than the control angle αl when the current is interrupted. This shows that when (α1-α2) is the deviation angle θ, the control angle is controlled to a value smaller than the actual control angle when the current is interrupted.

Therefore, even if the control is performed using the linear relationship of equation (3), the relationship between the DC voltage Va and the phase control signal VB becomes nonlinear.

On the other hand, looking at the relationship between the phase control signal VR and the DC current (average value) 1, the DC current Id can be expressed as shown in the following equation.

R: Equivalent resistance of the motor Substituting equation (3) into equation (4), the signal VR and DC current ■
The relationship between , and is calculated as follows.

As is clear from equation (5), when the current is continuous, the signal VR
The relationship between the current and the current is linear, or when the current is intermittent, the average voltage V, i becomes the induced voltage VM during the no-current period as described above. Since the converter 3 is in a disconnected state, the DC current uk Ia becomes small. Therefore, when the current is interrupted, the relationship between the signal VR and the current 1 becomes non-linear.

This relationship is shown in Fig. 6, where the continuous current range is a straight line9, and the intermittent current range is a curved line (Ca), (b).
, (C).

The characteristics (a), (b), and (C) change depending on the magnitude of the induced voltage VM. This means that when the induced voltage vM is large as shown in FIG. 7, the control angle α is made small in order to increase the average voltage vd, and when the induced voltage VM is small, the control angle is made large. This red instantaneous voltage vO is as shown in Fig. 7, but the ripple of the instantaneous voltage v, ) when the control angle α3 is small as shown in Fig. 7 (a) is shown in Fig. 7 (b). The control angle α4 (C4〉C
3), the ripple of the instantaneous lid pressure VQ becomes large. This is because, as the induced wake-up pressure VM becomes smaller, the average current for the current to continue flowing during the construction period increases.

As described above, as shown in FIG. 6, ttfljt(l command VR
The gain to the current T decreases rapidly when the current is intermittent.

By the way, as shown in FIG. 5, there is a difference between the control angle α1 that generates the DC average voltage when the current is interrupted and the ft1lJ angle α2 that generates the same DC average voltage Vdz when the current is continuous. Find out what the relationship is.

The deviation angle θ is expressed by the following formula.

θ=ψ−−−δl River...River (6) Also, δ in equation (6) is the flow angle, which is the value equal to the following equation, 9.However, it is valid only when θ is positive. When θ is negative, it is zero.

In equation (7), E2+, ω, and L are one pair,
After all, the deviation angle θ differs depending on the current ■1 and the control angle α, and the 8th
The percentage is shown in the figure.

That is, the DC average voltage Vd is expressed by the following equation.

, 3Vj V a 〒p, 2 CO3((1-θ) −
--- Old - (81 Therefore, when the current is intermittent, the actual control angle α' is the sum of the angle α of 71i11 and the deviation angle θ corresponding to the direct current ■1 and the control angle α. The uniform pressure is 3v'T Va = g2ωS (α'-θ) V7 = -E2CO3 (α+θ-θ) V2 = -E 2 CO5α ...
...(9) π and the relationship between the phase command signal VR and the average 'voltage vd is (
2) It becomes the same as Eq. That is, the gain characteristics when the current is intermittent can be made the same as when the current is continuous.

Based on the above theory, it is clear that nonlinear compensation can be achieved by adding the control deviation angle θ when the stain flow is interrupted.

Now, returning to FIG. 2, the operation will be explained.

The boundary value passage detector 30 turns on the switch 21 during uterine operation. In this state, function generator 14 is set as follows.

In FIG. 8, the current value kI at which the deviation angle θ becomes zero at each control angle α is set to values A, B, C, and D, so that the current value k1 at the deviation angle θ′ is set. The values of A' and B'.

Between c/ and D/ there is D/A two D'ri/A', D/B
=D'/B', and D/C=D'/C'.

Therefore, when the control angles are different, the coefficients corresponding to the control angle α, the coefficients of the tweezers D/A, D/B, and D/C are used as the current value kI.
, it is possible to match the %note at any control angle to the characteristic of α; 90°. That is,
The coefficient for the control angle α is 1'fr: As shown in Figure 3, the current value is k■ The deviation angle θ at any control angle can be determined by the function .

Now, the current value kI that is cut off below α=400,
When I was driving in 20, 06, I tried to solve the case of α2300°45°. α=300 and @ is &! As shown in Figure 3, Kkl = 2, so klki, = 0, 06
2 = 0.12, so the deviation angle θ is equal to 0.12.

On the other hand, when α=45°, it is 1.42 and kl
J,=0.085. Therefore, as shown in FIG. 4, θ=0.6 and α'=45.6°.

In addition, the control angle is intermittent when the control angle is 300 or less.
, = 0.045, then α = 30
When 1 = 2, klk l A = 0.0
4.5 X2 = 0.09. In this case as well, the deviation angle θ is the same. When α=45°, kl=1.4
2, so klkl = 0.0639 and θ2 3.3°
becomes. Therefore, the corrected control angle α' obtained from the adder 15 is α'=α230° when α=30°, and θ23.3° when α245°. α′= 4
It becomes 8.3°.

In this way, the set control angle α and the deviation angle θ with respect to the current are
By calculating the corrected control angle α' as α'=α10, the deviation angle θ that occurs during intermittent operation is added in advance to calculate the average value of the voltage in the hunting part in Figure 2 (a). Since the voltage is attempted to be lowered by '', the increase in pressure that occurs during intermittent operation is canceled out, resulting in an average blood pressure that corresponds to the set control angle α. In other words, since the characteristics are the same even during intermittent operation, it is possible to eliminate response deterioration.

Next, the setting of the royal circuit constant during periodic horizontal placement during trial operation will be explained with reference to FIG. 8'fd:.

When setting, drive the DC motor and switch 21.
Turn off.

Since the switch 21 is turned off, the deviation angle θ is not added to the set control angle α. When the angle selector switch 22 is turned on to side a, the values of i and i are output as shown in FIG. 4 when the correction surface of the correction angle calculator 14 becomes zero. In this state, the operation is performed with full current flowing until the direct current becomes continuous, that is, from the intermittent state to the continuous state as shown in Figure 9 (b).At this time, it is determined whether the current is intermittent or continuous. In order to detect the instantaneous current shown in Fig. 9 0 at the point where the ignition timing signal shown in Fig. 9 is generated, the instantaneous DC detector 2
7 is detected. The detected value is as shown in Figure 9 (Q).

If this detection □ signal is detected, the current zero detector 28 detects a detected value of I.
Detects less than or equal to o (a value very close to zero).

The detection signal of the current zero detector 28 is as shown in FIG. 9 (0). The rising edge of this signal is detected by the disconnection detector 29 as shown in FIG. The physical change point passage memory circuit 30 is turned on, and the switch 21 is turned on.
Close the 9J switch 22 to b 9111. On the other hand, at the disconnection change point, the input of the correction angle calculator 14 is 1.
In ■, the selector switch 22 is turned on to the a side, so the output of the split Naz5 is A+A=1, and the main circuit constant k
was the initial first shift.

Since the multiplier 24 multiplies by 1 and 1, the initial setting values are fi, k, /, and .Ia'. This value 1', l:,' is stored in the memory circuit 26 by the output signal of the disconnection detector 29. At the same time, selector switch 2
2 is turned on to the b side, the value A of the WO calculator 25 is the value at which the correction angle θ becomes zero, that is, when the main circuit constant k is set to '' where the main circuit constant matches the actual main circuit constant. It is a value. Therefore, the matching value of adder 24 is automatically adjusted.

As described above, according to the present invention, the main circuit power number is automatically set to the optimum value, so the positive trend region due to overcompensation, which was a drawback of the conventional method, does not occur, and the settings are adjusted according to the motor. Optimal nonlinear compensation can be performed without any need for deformability.

FIG. 10 shows a block diagram composed of digital circuits.

41 is the average value of current 11, instantaneous value! The voltage detector 2 detects the positive voltage of the power source by converting all analog to digital; the voltage detector 2 detects the speed of the pulse train signal of the 43-digit PLG; the gate pulse generator generates phase pulses according to the control angle finger α; A speed detector, 44 a processor that performs automatic setting (auto-tuning) calculations such as speed control calculations, current control calculations, control angle correction calculations, etc., 45 a memory for storing programs and data, and 46 a speed command etc. from a host controller. This is an interface circuit that answers input or actual RII flow and speed to a higher-level controller.

FIG. 11 is a control flow diagram of the nonlinear complementary fjf calculation part with autotuning, which is a feature of the present invention. First, all the control angle commands α(n) are calculated by current control calculation, and the average value of the current is 1. (rl) and v￥hour+1ti i, (
n) Detect k. Next, from the calculated ice formation α(n), the non-winning type correction factor 1kt(n) is calculated using the table shown in FIG. Here, it is determined whether or not the tuning mode is selected. If the tuning mode is selected, the horizontal axis in FIG. 4 is kl.

If 1 is used as the preset main circuit constant setting value, then 1kIl2kl(n)・k,・■,(n) are used. Here, if the magnetic current instantaneous value i, (n) is less than or equal to a preset intermittent current value, the coefficient is as follows. In other words, the setting is p. Further, the output control angle αω) is calculated by current control calculation and is set as the control angle α(n). Next, the instantaneous current i, (n) is i. If it passes, let it be the coefficient. Here, since the value A is the intermittent limit value, k can be a coefficient that matches the actual main circuit constant.

Once the constant k is set, the next process is to switch from the tuning mode to the driving mode. (1-
1) Calculate the correction angle θ(n) using the value of k calculated in the tuning mode, i.e., the calculated value, using the table J in Figure 8, and add it to the control angle αω) calculated by the current control calculation. , the output control angle α(n) is α(n) − α(n) + θ(n)
shall be. By doing so, when the intermittent region is entered, the correction is automatically activated, and optimal compensation can be performed.

As explained above, the present invention V 7t can be carried out without considering the royal circuit circuit 1t=r"1jjl jsugiiN!
I can almost say fjll.

[Brief explanation of drawings]

Fig. 1 is a block diagram of one row of static Leonard equipment, Fig. 2 is a block diagram showing an embodiment of the present invention, and Fig. 3 is a correction coefficient calculation diagram.
! , the characteristic diagram of the device, Figure 4 is the characteristic diagram of the correction angle calculator, and Figure 5 is the characteristic diagram of the correction angle calculator.
Figure 6 shows the voltage DC waveforms during intermittent and continuous current, Figure 6 shows the characteristics of the phase control signal and motor magnetic current, Figure 7 shows the cardiac pressure and cardiac current waveforms, and Figure 8 shows each node.・Power transformer, 2...
AC current transformer, 3... Thyristor converter, 4... DC core, M view, 5... Speed detector, 9... Automatic pulse phase shifter, 8... Current control circuit, 6 ...speed control circuit,
7... Current detector, 10... Inverse cosine variable Houhe, 11.
...Correction coefficient calculator, 12.13...J+f) 1 unit, 14...Correction angle calculator, 15...Adder, 21.
...Switch, 22...Selector switch, 26...Memory circuit, 27...Instantaneous current detector, 28...Zero current detector, 29...Disconnection change point detector, 30...・Change point passage record・ is 468 praise 1 mouth SR 3rd d) Lih-pβ1 Q (41β・Iα pull 5 mouth □ ￥ saliva￥ -'7 帛 (α) (Inn 3rd) 1 Ka・Lo .House 10th 11th

Claims (1)

[Claims] 1. A two-ristor converter that supplies power to a load whose back electromotive force changes in magnitude, a current control means that outputs a phase control signal according to a deviation between a current command signal and a current detection signal of the load current, and the phase control The control angle calculation means for determining the set control angle based on the signal and the converter must generate the same DC average voltage when the load current is continuous and when the load current is intermittent. 1iIII deviation angle calculating means for calculating the control deviation angle from the set control angle and the load current; current disconnection detection means for detecting the load current and disconnection boundary value; inputting the set value of the main circuit constant; and constant adjustment means for correcting the setting value of the main circuit constant according to the magnitude of the control deviation angle at the time of value detection, and the deviation angle calculation means calculates the control deviation angle based on the corrected main circuit constant. 1. A control device for a thyristor converter, characterized in that the magnitude of the load current is corrected when the load current is applied.